The main wing of an aircraft has so far been produced using an aluminum alloy or a titanium alloy, but recently, has used a carbon fiber-reinforced plastic (CFRP) comprising a combination of carbon fibers and resin.
In the aircraft, damage to the wing poses a danger directly leading to a serious accident. Thus, it is very important to perform operations for maintenance and checkup, thereby finding a defect.
Defects occurring in the wing of the aircraft include a flaw, foreign matter, a peel, and so on. As a nondestructive examination for detecting these defects, an ultrasonic flaw detection test or inspection is adopted.
A general technique for the ultrasonic flaw detection test will be explained first of all.
If a flaw detection range A is broad, as shown in FIG. 10, a flaw detection path P is established or set. Along this flaw detection path P, an ultrasonic probe is moved to acquire a flaw detection waveform signal (response waveform signal) by an ultrasonic wave.
If, at a certain point on the flaw detection path P, this site is a sound site, a flaw detection waveform signal, which comprises a surface echo Es and a bottom echo Eb as shown in FIG. 11, is obtained. If, at a certain point on the flaw detection path P, this site is a defective site, a flaw detection waveform signal, which comprises the surface echo Es, the bottom echo Eb, and a defect echo Ed as shown in FIG. 12, is obtained.
If the flaw detection range A is narrow, it is possible to detect a defect (detect the defect echo Ed) by visually observing the waveform of the flaw detection waveform signal throughout the flaw detection path P. If the flaw detection range A is wide, on the other hand, the visual confirmation of the waveform of the flaw detection waveform signal over the entire flaw detection path P takes a huge time.
Under these circumstances, the flaw detection waveform signal is subjected to signal transformation or signal conversion to convert this signal into a flaw detection image signal representing a flaw detection image in which the luminance or brightness value of the defective site changes relative to the brightness value of the sound site. By visually confirming the flaw detection image based on this flaw detection image signal to determine whether there is a defect or not, a flaw detection examination or test is conducted in a short time.
If the presence of the defect is observed, a detailed analysis can be made by confirming the flaw detection waveform which is the basis for the flaw detection image.
As techniques for signal transformation for converting the flaw detection waveform signal into the flaw detection image signal, TOF image signal transformation and AMP image signal transformation are known.
To carry out signal transformation, a gate G with a predetermined range and at a predetermined height is set at a position between the surface echo Es and the bottom echo Eb in each of FIG. 13 showing the flaw detection waveform signal on the sound site and FIG. 14 showing the flaw detection waveform signal on the defective site.
For conversion into a TOF image, the position where the set gate G and the defect echo Ed intersect is converted into a brightness value. For conversion into an AMP image, the height of the defect echo Ed intersecting the set gate G is converted into a brightness value.
If a plurality of the defect echoes are present, there are versions, for example, in which the first echo that intersects the gate G is used as the defect echo Ed, or the highest echo of the echoes intersecting the gate G is used as the defect echo Ed.
FIG. 15 shows an example of the flaw detection image of the sound site. FIG. 16 shows an example of the flaw detection image (TOF image) of the defective site. FIG. 17 shows an example of the flaw detection image (AMP image) of the defective site.
As shown in these drawings, the flaw detection image shows the brightness of the defective site changing relative to the brightness of the sound site, and enables the defective site to be inspected visually with rapidity and ease.
The waveform shape of the flaw detection waveform signal and the status of occurrence of a noise differ according to the type of a defect or the plate thickness, shape, and material of the flaw detection site. Thus, a single flaw detection image may be insufficient to determine accurately whether there is a defect or not.
Thus, a plurality of gates different in position or height are set, in accordance with the type of the defect or the plate thickness, shape, and material of the flaw detection site, to generate a plurality of flaw detection image signals, and a plurality of flaw detection images by the plurality of flaw detection image signals are visually confirmed, whereby the presence or absence of the defect is determined overall.
That is, one flaw detection waveform signal obtained when the probe is moved along the flaw detection path P is subjected to signal transformation by a plurality of different signal transformation techniques to obtain a plurality of flaw detection image signals, and flaw detection images by these plural flaw detection image signals are used to judge globally whether the defect is present or not.
Conversion of a flaw detection waveform signal into a TOF flaw detection image signal, and conversion of a flaw detection waveform signal into an AMP flaw detection image signal are different techniques for signal transformation. Moreover, the conversion of a flaw detection waveform signal into a TOF flaw detection image signal, with the gate G differing, is also a different technique for signal transformation. Furthermore, the conversion of a flaw detection waveform signal into an AMP flaw detection image signal, with the gate G differing, is also a different technique for signal transformation. In these senses, these methods are called “a plurality of different signal transformation techniques”.
A general example of globally judging the presence or absence of the defect by the flaw detection images by the plurality of flaw detection image signals obtained by the signal transformation of the one flaw detection waveform signal will be described below.
If the types of the defect are two types, D1 and D2, the plate thickness of the flaw detection site is 10 mm, the materials of the flaw detection site are of two types, M1 and M2, and the shape of the flaw detection site is of two types, F1 and F2, for example, a flaw detection image I1 by a TOF flaw detection image signal obtained by signal transformation of the flaw detection waveform signal with a first gate being set, a flaw detection image I2 by a TOF flaw detection image signal obtained by signal transformation of the flaw detection waveform signal with a second gate being set, a flaw detection image I3 by an AMP flaw detection image signal obtained by signal transformation of the flaw detection waveform signal with a third gate being set, and a flaw detection image I4 by an AMP flaw detection image signal obtained by signal transformation of the flaw detection waveform signal with a fourth gate being set can be used for defect evaluation.
If the allocation of the flaw detection images I1 to I4 is made as in Table 1, the flaw detection images I1 to I4 are visually confirmed in sequence, whereby all the defects can be evaluated.
TABLE 1Example of allocation for defect evaluation by flaw detection imagesPlate thicknessMaterialFlawType ofof flawof flawShape of flawdetectiondefectdetection sitedetection sitedetection siteimageD110M1F1I1D110M1F2I1D110M2F1I2, I3D110M2F2I4D210M1F1I1D210M1F2I2D210M2F1I1D210M2F2I2, I4
A conventional technique for conducting the ultrasonic flaw detection test of the main wing of an aircraft will be explained.
With the test of the main wing of the aircraft, the whole surface of the main wing is ultrasonically tested for flaw detection. However, it takes too much time to analyze all of flaw detection waveforms by a flaw detection waveform signal obtained by this test. Thus, the acquired flaw detection waveform signal is subjected to signal transformation by a plurality of different signal transformation techniques to be converted into a plurality of flaw detection image signals. On a plurality of flaw detection images by the plurality of flaw detection image signals, a defective site is identified visually.
On the flaw detection image, the brightness value of the defective site is increased (or decreased) compared with a sound site. By visually confirming a change in brightness between the sound site and the defective site, therefore, the defective site is identified.
In this case, the type and arrangement of the flaw detection image to be visually confirmed differ according to the type of the defect to be detected, or the difference of the site (shape, plate thickness, material). Moreover, the display magnification and contrast of the flaw detection image in which the defect is easy to detect are also different. This is because a manner of reflection of an ultrasonic wave differs according to the type of the defect or the plate thickness, shape or material of each site.
Thus, an inspector repeatedly changes the type and arrangement, display range (magnification) and contrast of the flaw detection image displayed, according to the type of the defect and the difference of the site, and detects the defect by so doing.
The details of such a conventional inspection technique will be described by reference to FIG. 18 which is a flowchart, and FIGS. 19 to 21 showing flaw detection waveform images displayed on the display screen of an image display device.
Assume that a plurality of flaw detection image signals created by signal transformation of a flaw detection waveform signal, which was obtained by the ultrasonic flaw detection test of the main wing of an aircraft, by a plurality of different signal transformation techniques have been previously stored on a flaw detection image signal database.
In starting visual inspection work (Step S1 in FIG. 18), an operator opens a plurality of flaw detection images I1 to In derived from a plurality of flaw detection image signals (Step S2).
Then, the type of a defect to be evaluated (for example, the defect D1 of the first type) is determined (Step S3).
What type of defect should be investigated is judged based on knowledge possessed by the operator (knowledge of the structure, shape, plate thickness and material of the wing, knowledge of the characteristics of the ultrasonic flaw detection test, knowledge acquired through inspection experiences accumulated, and so on).
The operator selects one or a plurality of the flaw detection images required, according to the type of the defect determined, and displays the selected image or images in an aligned arrangement on a display screen 1 of an image display device as shown in FIG. 19 (Step S4).
If the type of the defect is D1, for example, the flaw detection image I1 and the flaw detection image I3 are selected, and both of the flaw detection images I1 and I3 are displayed in an side-by-side arrangement.
In this case, the flaw detection images I1 and I3 are each displayed in such a manner as to be superposed on, for example, partitioned regions R1 to R28 separated into 28 (4×7) regions by parting lines indicated by short dashed lines in FIG. 19. That is, the flaw detection images I1 and I3 are each divided into images on the partitioned regions R1 to R28.
Which of the flaw detection images should be selected according to the type of the defect, and how the selected images should be arranged, are judged based on the knowledge possessed by the operator (knowledge of the structure, shape, plate thickness and material of the wing, knowledge of the type of the defect, knowledge of the characteristics of the ultrasonic flaw detection test, knowledge acquired through inspection experiences accumulated thus far, and so on).
The operator changes the display range of the flaw detection images I1, I3.
Concretely, the starting position of X and Y coordinates displayed, and the ending position of the X and Y coordinates are changed to specify the partitioned regions to be displayed. Also, the display magnification is optimally changed to display, on an enlarged scale, an image in a predetermined range, for example, on the partitioned region R1 of the flaw detection images I1 and I3 (Step S5), as shown in FIG. 19.
To what degree the display magnification should be adjusted to be increased, and the image on which of the partitioned regions should be displayed, according to the type of the defect, are judged based on the knowledge possessed by the operator (knowledge of the structure, shape, plate thickness and material of the wing, knowledge of the type of the defect, knowledge of the characteristics of the ultrasonic flaw detection test, knowledge acquired through inspection experiences accumulated thus far, and so on).
In the state of FIG. 20, the image on the partitioned region R1 has merely been enlarged. Hence, the contrast between an image showing the defective site (a black-colored portion) and the sound site is not clear.
Thus, the operator makes a contrast adjustment for the enlarged images of the flaw detection images I1, I3 on the partitioned region R1 (FIG. 20) to enhance the contrast between the image of the sound site and the image of the defective site (Step S6).
To what degree the contrast should be adjusted, according to the type of the defect, is judged based on the knowledge possessed by the operator (knowledge of the structure, shape, plate thickness and material of the wing, knowledge of the type of the defect, knowledge of the characteristics of the ultrasonic flaw detection test, knowledge acquired through inspection experiences accumulated thus far, and so on).
The operator visually confirms the images (FIG. 21) obtained by enlarging, and adjusting the contrast of, the flaw detection images I1, I3 on the partitioned region R, to evaluate whether the defective site is present or not (Step S7).
The operator changes the display range of the flaw detection images I1, I3 from the image enlarged from the image on the partitioned region R1, successively, to the images enlarged from the images on the partitioned regions R2 to R28. When selecting the enlarged image on each of the partitioned regions R2 to R28, the operator performs the optimal changing of the display magnification, the contrast adjustment, and the visual inspection of the defective site, which are shown in Steps S5, S6 and S7, respectively, in accordance with the situation of the selected partitioned region.
After all the images on the partitioned regions R1 to R28 are inspected for the presence or absence of the defect D1 (Step S8), the type of the defect to be evaluated is changed to the defect of a different type (for example, defect D2) (Step S3).
For the defect D2, an inspection of the defective site is conducted in the same manner as for the defect D1.
In this manner, the type of the defect is sequentially changed, and after the inspection of the defective site is conducted for all types of defects, evaluation is completed (Step S9). Then, analysis work on the flaw detection images is completed (Step S10).
Depending on the type of the defect, the analysis of the flaw detection image in a specific partitioned region may be skipped, and this analysis may be made in a next partitioned region.
The reason is as follows: Knowledge or experience may teach, from the beginning, that depending on a specific image selected and a specific defect selected, defect detection in a specific partitioned region is impossible. In this case, the analysis of the flaw detection image in the specific partitioned region is skipped in order to cut down on time and labor.